Allying topology and shape optimization through machine learning algorithms

Structural optimization is part of the mechanical engineering field and, in most cases, tries to minimize overall weight a given design domain, subjected functionality constraints terms stresses or displacements. The relevant techniques are topology and shape optimization. Topology provides optimal material distribution layout into given, static, domain. On other hand, combination parameters that define required parametrization domain's boundary. Both have strengths weaknesses, thus hybrid approach combines former will more general structural framework take advantage their synergistic combination. difficulty arises when communicating both for which, this paper, we propose machine learning-based methodology.